Parallel Intelligence in Semantic Digital Twins: An Interactive Decision-Support System for Indoor Comfort

Abstract

While buildings should be designed for their occupants, many buildings fail to satisfy their expectations. The growing research body on personal comfort models is promising to reduce the gap between perceived and predicted comfort; however, measuring perceived comfort levels, integrating them with other heterogeneous information, and making decisions based on the integrated data is a challenge. This paper combines semantic web technologies with an interactive dashboard to measure and integrate the occupants' feedback on indoor comfort. A personal comfort model then calculates an individual's preferred indoor climate. The system is tested in a case study with two occupants and shows that the digital twin can use the human-machine interaction to improve decision-making.